Let me be direct: we are not having a theoretical conversation about AI and work anymore. That ship sailed. The question now is whether you're paying attention to what's actually happening — or still waiting for someone to tell you when to panic.

I build things. I've founded an AI platform, run marketing for a Web3 academy, built a real estate company in Nigeria. I move fast by necessity. So when I say AI is already reshaping how I work, I don't mean I'm experimenting with it. I mean it's changed what I hire for, how I write, how I make decisions, and what I consider a competitive advantage.

Here's my read on where we actually are.

The replacement isn't coming. It's already happened.

The debate about "will AI take jobs" is stale because it assumes a dramatic, visible shift — mass layoffs, robots on assembly lines, an obvious before and after. That's not how it's working. It's quieter. A company that used to hire three content writers now hires one, with AI handling the first drafts. A startup that needed a data analyst on staff now uses Claude or GPT to run queries and synthesize reports. The headcount never goes up. Nobody gets fired loudly. The job just doesn't get created in the first place.

That's the displacement. It's in the unfilled roles, not the pink slips.

And the roles most exposed aren't the ones people expect. It's not just repetitive tasks. It's mid-level knowledge work — the kind that pays $60K–$100K and requires a degree but not deep expertise. First-pass research. Standard copywriting. Basic legal summaries. Junior financial analysis. These are getting compressed fast.

What actually survives

I'll tell you what I've noticed holding up: judgment, relationships, and context that can't be replicated without lived experience.

Judgment means knowing when to trust the output and when to push back. Most people using AI right now have no instinct for when it's confidently wrong. That's a skill. It's also a moat — for now.

Relationships are still deeply human. Sales, partnerships, leadership — these depend on trust built over time between people. AI can assist the process. It can't own it.

Context is underrated. The person who understands the culture of a company, the political dynamics of a market, the specific way a customer segment thinks — that person has something an LLM can't easily replicate. AI is trained on the general. You have access to the specific.

What doesn't survive: being the person whose main value is executing a predictable task at scale. Research without synthesis. Writing without voice. Analysis without recommendation. If the job is to produce an artifact that someone else will then think about — that's the exposure zone.

The smart move right now

I see two types of professionals right now. The first is waiting to see how things develop. The second is rebuilding their value proposition around what AI can't do — and using AI aggressively to amplify everything else.

The second group is winning.

Concretely, here's what I think the smart move looks like:

Get fluent, not just familiar. There's a difference between occasionally using ChatGPT and actually understanding how to prompt well, where these tools break, and how to integrate them into a real workflow. Fluency is the new baseline. Familiarity is table stakes.

Compress low-value work ruthlessly. Any task that is repeatable, research-heavy, or draft-based — AI should be doing the first 80%. Your job is the last 20%: judgment, refinement, positioning. If you're still spending four hours writing a first draft, you're leaving leverage on the table.

Get specific about your edge. Generalists are getting squeezed. Not because generalism is worthless, but because AI is a very good generalist. The question you need to answer is: what do I know or see or do that's harder to replicate? Build more of that.

Don't oversell your adaptability. Show it. I've seen a lot of people add "AI-powered" to their LinkedIn bios without changing anything about how they work. That's noise. What matters is whether you're actually building in public, shipping faster, making better decisions because of the tools available to you. Show, don't tell.

What I'm actually watching

The thing that keeps me alert isn't the headline stuff. It's the speed of capability improvement. A year ago, I was skeptical about AI handling nuanced strategic work. I'm less skeptical now. That trajectory matters.

The people who will get caught off guard aren't those who dismissed AI entirely — they're people who adopted it just enough to feel safe, without going deep enough to see what's coming next.

I'm not interested in being part of that group. I don't think you are either, or you wouldn't be reading this.

The work is changing. Your job is to change faster.